Heterobifunctional Ligase Recruiters Enable pan-Degradation of Inhibitor of Apoptosis Proteins

Proteolysis targeting chimeras (PROTACs) represent a new pharmacological modality to inactivate disease-causing proteins. PROTACs operate via recruiting E3 ubiquitin ligases, which enable the transfer of ubiquitin tags onto their target proteins, leading to proteasomal degradation. However, several E3 ligases are validated pharmacological targets themselves, of which inhibitor of apoptosis (IAP) proteins are considered druggable in cancer. Here, we report three series of heterobifunctional PROTACs, which consist of an IAP antagonist linked to either von Hippel-Lindau- or cereblon-recruiting ligands. Hijacking E3 ligases against each other led to potent, rapid, and preferential depletion of cellular IAPs. In addition, these compounds caused complete X-chromosome-linked IAP knockdown, which was rarely observed for monovalent and homobivalent IAP antagonists. In cellular assays, hit degrader 9 outperformed antagonists and showed potent inhibition of cancer cell viability. The hetero-PROTACs disclosed herein are valuable tools to facilitate studies of the biological roles of IAPs and will stimulate further efforts toward E3-targeting therapies.


■ INTRODUCTION
In the last decade, significant advancements have been made in the field of proteolysis targeting chimeras (PROTACs).−6 Classical PROTACs are heterobifunctional compounds comprising a ligand that binds to a target protein of interest, a ligand that binds to and recruits an E3 ubiquitin ligase, and a linker tether.Such molecules can facilitate the formation of a ternary target−PROTAC−E3 ligase complex, followed by ubiquitination of the target protein and its subsequent degradation by the proteasome. 7ROTACs possess several advantages over conventional inhibitors as they exert their action catalytically, resulting in a potent intracellular degradation of the target proteins.Moreover, PROTACs can discriminate between similar proteins within the same family or even protein isoforms, thus allowing exclusive target-selective degradation. 8,93 ubiquitin ligases orchestrate protein turnover via facilitating substrate proximity and ubiquitin transfer.They encompass a diverse group of more than 600 enzymes, with most E3 ligases belonging to the really interesting new gene (RING) family.Many have crucial roles in various biological processes 10 but are also implicated in multiple diseases.−15 Cellular RING E3 ligases are large multi-subunit complexes but usually do not possess a defined ligand-binding site rendering the inhibitor design difficult.Nevertheless, traditional approaches yielded potent compounds targeting murine double minute 2 (MDM2), von Hippel-Lindau (VHL), and inhibitor of apoptosis (IAP) proteins. 11The consequence of binding to these E3 ligases is disrupting protein−protein interactions between ligases and their respective substrates.E3 ligases can also be degraded via proximity-induced ubiquitination.−19 Directing different E3 ligases against each other by heterodimeric PROTACs also proved to be a productive strategy for their depletion.We recently reported preferential degradation of cereblon (CRBN) over VHL with molecules assembled from pomalidomide-based CRBN binders and a VHL ligand (CRBN-6-5-5-VHL, Figure 1A). 20The prevalence of VHL over CRBN was also observed in a separate study by Ciulli and colleagues (14a, Figure 1A). 21On the contrary, linking MDM2 inhibitors to lenalidomide resulted in MDM2 degradation (MD-224 and PROTAC 32, Figure 1A). 22,23Of note, CRBN levels were not monitored for the latter two examples, thereby not entirely confirming the unilateral degradation of MDM2.Recently, the compendium of heterobifunctional ligase degraders was extended by KEAP1-CRBN recruiters (PROTAC 14 and NJH-04-087, Figure 1A) that preferentially degrade KEAP1. 24,25ellular IAP1 (cIAP1/BIRC2), cellular IAP2 (cIAP2/ BIRC3), and X-chromosome-linked IAP (XIAP/BIRC4) have been studied in great detail because of their critical role in controlling the apoptotic machinery. 26−29 This clinical significance has translated to the development of numerous mimetics of the IAP-binding motif (i.e., the N-terminal Ala-Val-Pro-Ile moiety) 30−32 of the second mitochondria-derived activator of caspases (SMAC), which functions as an endogenous IAP antagonist. 33,34Several SMAC-mimicking IAP monovalent and bivalent antagonists have entered into clinical trials for the treatment of various cancers (Figure S1). 35,36However, they demonstrated low efficacy as single agents, and current clinical evaluations are limited to combination studies with other cytotoxic drugs, radiation, and immunotherapy. 37AP antagonists have profound effects on cIAPs levels.Both monovalent and bivalent SMAC mimetics bind to the baculoviral IAP repeat (BIR) type 3 domain of IAPs.This event promotes a rapid and ubiquitin-and proteasomedependent loss of cIAP1 and cIAP2.38,39 It was suggested that BIR3-binding IAP antagonists destabilize a closed/  20,21 whereas MD-224 and PROTAC 32 cause depletion of MDM2 via CRBN-mediated ubiquitination.18,23 PROTAC 14 and NJH-04-087 exhibited degradation of KEAP1 over CRBN.24,25 (B) Schematic representation of the designed compounds in this work.
autoinhibited form of cIAP (by blocking the crucial BIR3-RING domain interactions), resulting in dimerization, E3 ubiquitin ligase activation, autoubiquitination, and proteasomal degradation. 40,41Although most IAP antagonists do also possess an affinity for the BIR3 domain of XIAP, this rarely results in XIAP autodegradation. 38IAP antagonists such as bestatin and more recently the improved LCL161 also represent valuable E3-recruiter moieties for the assembly of heterobifunctional protein degraders. 42ncouraged by our successful outcomes with heterodimerizing PROTACs and the fact that IAPs are validated anti-cancer targets, we decided to include IAPs in the hetero-PROTAC approach for E3 modulation (Figure 1B).We systematically designed three series of bifunctional molecules.Two series were assembled by linking the selected IAP ligand 43 to two VHL ligands with differently oriented exit vectors.In the third series, the CRBN ligand pomalidomide was incorporated into hetero-PROTACs (for E3 ligand structures, see Table S1).Our efforts to induce IAP degradation through hijacking another E3 ligase via hetero-PROTACs led to selective depletion of IAPs, including XIAP, and, in one case, also to selective XIAP degradation.Notably, the pan-IAP deficit could translate into very potent inhibitors of cancer cell viability, thus further substantiating the rationale of our strategy.

■ RESULTS AND DISCUSSION
Design and Synthesis of Bifunctional IAP PROTACs.It is widely accepted that linkers play a significant role in PROTAC activity, as even subtle differences in length or composition influence the degradation activity and selectivity. 44,45Three libraries of heterobifunctional PROTACs were systematically designed with eight linkers of varying length and chemical composition in each library (Tables S2−S6).The synthesis of the first series of IAP-VHL hetero-PROTACs (Series 1) was accomplished by a straightforward approach employing chloro to carboxylic acid (Cl-to-CO 2 H) linkers L1a−L8a (Table S2).Most of the linkers were acquired by a BAIB/TEMPO-mediated oxidation of the appropriate primary alcohol precursors.In contrast, L6a, L7a, and L8a were prepared by elongating C6−O−C6 or C6−O−C5 alcohols with tert-butyl bromoacetate, tert-butyl 5-bromopentanoate, or tert-butyl 6-bromohexanoate (Schemes 1 and S1), respectively, followed by deprotection of the tert-butyl ester.Linkers L2a− L8a were first coupled to the VHL ligand VH032 (VHL1 ligand, 68), 46 and the obtained conjugates 75−81 were applied to alkylate the Boc-protected IAP ligand 65 (Scheme 2A).Because intramolecular cyclization occurred when coupling the C5 linker L1a to the VHL1 ligand, L1a was first attached to the Boc-protected IAP ligand 65 or to the negative control IAP ligand 66.This was followed by deprotection of the terminal carboxylic acid before coupling with either VH032 (68) or a methylated VHL ligand (Me-VHL1, 69) which is known to enhance the VHL-binding affinity. 47The desired hetero-PROTACs 1−9 and controls 10a and 11 were obtained following Boc deprotection of the corresponding IAP ligands (Scheme 2A).Additional control compounds were obtained by derivatization of PROTAC 9 into dimethyl-(10b) or acetyl-(10c) bearing analogues.
For the second IAP-VHL series (Series 2), a different exit vector at the VHL side was employed, presumably leading to differently oriented ternary complexes. 48,49For this subseries, methane-sulfonate to chloro (OMs-to-Cl) linkers L1b−L8b (Table S3) were utilized as crucial building blocks.Most of these (L1b−L5b) were prepared by subjecting the selected alcohols to mesylation.At the same time, L6b, L7b, and L8b were obtained through elongation of C6−O−C6 or C6−O− C5 mesylates with the corresponding tetrahydropyranyl (THP)-protected diols.Following cleavage of the THPprotecting group, mesylates were prepared (Schemes 1 and S5−S7).The obtained linkers were attached to the phenolic VHL ligand (VHL2 ligand, 71) through O-alkylation.These ligand−linker conjugates 83−90 were then connected to the IAP ligand 65, followed by removing the Boc-protecting group under acidic conditions to yield hetero-PROTACs 12−19 (Scheme 2B).
In the third series of hetero-PROTACs, OMs-to-Cl linkers were used to alkylate the IAP ligand 65, and the resulting chloro-linker-IAP ligand conjugates 91−98 were converted to amino-functionalized building blocks via azidolysis and hydrogenolysis.Finally, these amine building blocks 99−106 were reacted with 4-fluorothalidomide (72) in a nucleophilic aromatic substitution.Removal of the Boc protecting group yielded the envisioned IAP-CRBN heterobifunctional PRO-TACs 20−27.For the assembly of negative control compounds 28 and 29, an N-methylated thalidomide derivative 73 was used in place of 4-fluorothalidomide (72) (Scheme 3).
The physicochemical properties of all hetero-PROTACs (Tables S4−S6) and known IAP inhibitors (Table S7) are provided in the Supporting Information.Despite encompassing a wide range in terms of lipophilicity (elog D 7.4 from 3.4 to 5.8), a similar activity window was observed for IAP-VHL Series 1 PROTACs.In Series 2, the most lipophilic hetero-PROTAC 19 showed the lowest IAP degradation levels at 0.1 μM, whereas for IAP-CRBN series, high PROTAC lipophilicity led to XIAP-selective depletion (Figure 2 and Table S6).
The effect of representatives from Series 1 on IAP depletion was significantly enhanced in comparison to the incorporated IAP ligand CST530 itself (Figures 2, 3, and S2), which caused only potent cIAP1 autodegradation but moderate cIAP2 depletion in MM.1S cells.As expected, treatment with the VHL inhibitor VH298 did not affect the levels of IAPs.Hetero-PROTACs, on the other hand, induced complete cIAP1 and XIAP degradation even at 0.1 μM.In addition, a substantial reduction of cIAP2 levels was observed for these hetero-PROTACs (Figures 2 and S2−S4).We were particularly pleased with the ability of PROTACs to degrade XIAP, which was rarely down-regulated by monovalent or bivalent IAP ligands in all previous studies.Of note, after a 16 h treatment with 0.1 μM concentration, a reduction of VHL30 levels was observed for all Series 1 PROTACs (Figures 2 and S2).However, this effect was less pronounced for compounds containing long and hydrophobic linkers (i.e., PROTACs 6−  8).The most profound degradation of VHL30 (71% protein degradation, as quantified by western blot) was observed for hetero-PROTAC 3, meaning that both ligases were degraded simultaneously at 0.1 μM concentration.The effect on VHL30 levels was very similar at 1 μM (Figure S5).Bidirectional degradation was in stark contrast to the properties of our CRBN-VHL hetero-PROTAC CRBN-6-5-5-VHL, which selectively degraded CRBN. 20,50Similarly, the VHL-targeting homo-PROTAC CM11 only caused a reduction of the long VHL isoform but spared the 19 kDa protein. 17Comparative analysis between 1, 2, and CM11 confirmed similarities between PROTACs whereby IAP-VHL hetero-PROTAC 1 also notably and dose dependently reduced VHL19 levels (42% remaining VHL19 at 1 μM, Figure S6).We estimate that the effects of IAP-VHL degraders would not be masked by the hypoxia-inducible factor (HIF)-dependent hypoxic response because recent results showed that homo-PROTAC-mediated VHL30 degradation or siRNA-mediated knockdown of VHL leads to almost undetectable stabilization of HIF-1α. 17,51fter the initial PROTAC screening, 1 was selected for further optimization due to its comparatively small size and thus the higher chance to overcome PK/PD penalties. 52We modified the compound by incorporating the Me-VHL ligand 69 with improved binding affinity for VHL into the hetero-PROTAC.The resulting compound 9 (Figure 4A) showed enhanced pan-IAP degradation in MM.1S cells at even lower concentrations (Figure 4B).Interestingly, also stronger VHL19 degradation was observed at 1 μM (24% remaining VHL19).Densitometric quantifications of western blotting bands after treatment with hetero-PROTAC 9 in MM.1S cells revealed DC 50,16h values of 2.4 nM (cIAP1), 6.2 nM (cIAP2), and 0.7 nM (XIAP).Maximum cIAP1, cIAP2, and XIAP degradation (Dmax) of 99, 90, and 99%, respectively, at 0.1 μM concentration of 9 was achieved (Figure 4C).A head-tohead comparison of 9 with birinapant demonstrated that the latter caused more pronounced cIAP1 degradation, whereas 9 outperformed birinapant in depleting cIAP2 and XIAP (Figure S7A).On the other hand, AZD5582 showed stronger cIAPs degradation than IAP-VHL hetero-PROTAC 9 but did not influence XIAP levels even at 1 μM (Figure S7B).
Profiling the activities of the second series of hetero-PROTACs, where we utilized a different linker exit vector, revealed a degradation profile similar to that of the first series.Namely, hetero-PROTACs 12−14 with C5, C8, and C4−O− C4 linkers, respectively, induced the most potent pan-IAP degradation (Figures 2 and S3).Unidirectional ubiquitination between the two E3 ligases was again observed only for hetero-PROTACs with long linkers (compounds 17 −19).In line with this, also no effect on VHL19 levels was seen (Figures 2 and  S3).In terms of achieving degradation of a pair of IAPs, 19 seemed interesting as it caused dual cIAP1/XIAP degradation at 0.1 μM concentration in MM.1S.However, profiling of the To investigate the relative ability of E3 ligases to induce degradation of each other upon treatment with the IAP-CRBN hetero-PROTACs, MM.1S cells were used.For 21−26, consistent, unidirectional, and distinct degradation of all three IAPs was observed already at 0.1 μM concentration (Figure 2).Of these, compounds 22, 25, and 26 showed pronounced pan-IAP depletion, and, concurrently, they induced substantial IKZF3 degradation at 1 μM as an effect of modulating the substrate scope upon pomalidomide binding (Figure S4B).At 0.1 μM, only 22 caused depletion of IKZF3, with approximately 40% of IKZF3 degraded after 16 h treatment.This dual mode could be useful in settings where these secondary effects are desirable.The most intriguing finding within the IAP-CRBN hetero-PROTAC series was observed for compound 27, equipped with the longest linker.
An isoform-selective XIAP degradation was indicated after 16 h-treatment at 0.1 μM (Figures 2 and S4A).A significant and selective decrease of XIAP levels compared to cIAP1 and cIAP2 was confirmed on a proteome level, where MM.1S cells were treated with hetero-PROTAC 27 for 3 h (see Figure 6B).This result unveils that IAP selectivity within the IAP-CRBN hetero-PROTAC series can be tuned by linker modifications.
Mechanistic Considerations.To understand the mechanism of E3 recruitment and ubiquitin transfer, we tested a set of control compounds with inactivated IAP-or VHL-binding motifs (Table S4).As little was known about appropriately rendering IAP ligands inoperative, we synthesized a series of putative IAP-non-binding controls 10a−d (Scheme 2A and Figure S9).In 10a, the stereochemistry of the critical N-methyl alanine portion was inverted or substituted with a second methyl group (10b).However, literature data indicated remaining affinity for the XIAP-BIR3 domain. 53Indeed, 10a and 10b were still able to induce pan-IAP or cIAP1 and cIAP2 degradation, respectively (Figure S9).Further increasing the size of the N-terminal substituent and lowering the basicity in 10c (R = acetyl) and 10d (R = Boc) led to inactivated PROTACs.Both methylation and acetylation were performed through a convenient late-stage modification reactions of the final PROTAC 9, highlighting the general utility of these transformations to produce inactivated IAP-recruiting PRO-TACs.By analogy with series 10, hetero-PROTAC 11 (VHLent) possessing a VHL non-binding diastereomer only induced cIAP1 degradation (Figure 4B), which is a common attribute of IAP antagonists.Next, cellular activities of hetero-PROTACs 2, 4, and 6 were evaluated in chronic lymphocytic leukemia cells (HG3), for which a VHL-knockout cell line was created (Figure 5).In both VHL +/+ and VHL −/− cells, cIAP1 autodegradation was observed after treatment with our hetero-PROTACs, demonstrating that ligand binding is the conditio per quam.In contrast, recruitment of a VHL is required for XIAP degradation as this occurred only in HG3 wild-type cells.Thus, VHL knockout confirmed the involvement of E3 ubiquitin ligase CRL2 VHL in the induced degradation of XIAP (and, at least in part, cIAP1) by these IAP-VHL heterobifunctional PROTACs.This provides additional evidence that the degradation of IAPs relies on the formation of a hetero−ternary complex consisting of both ligases and the degraders.
When evaluating the time dependence of hetero-PROTAC 9, we observed complete degradation of cIAP1, cIAP2, and XIAP already after 3 h at 0.1 μM compound concentration.Interestingly, the effect on VHL30 depletion was most pronounced after 6 h of treatment in MM.1S cells (Figure S10A).
Next, we examined the persistence of IAP degradation in MM.1S cells after a single exposure to 1 μM of 9 and subsequent removal of the compound.Results indicated a nearly full and stable pan-degradation of IAPs up to 72 h.VHL19 and VHL30 levels restored more rapidly following drug washout (Figure S11A).Nevertheless, intracellularly cycling quantities of PROTAC that remain inside the cells after washout may be sufficient to generate these characteristics.In contrast, when the system was further challenged with the competing VH298 after the washout, XIAP levels increased more rapidly (Figure S11B), consistent with an IAP antagonist mode and an unleashed resynthesis of XIAP.In a series of experiments where individual IAPs were knocked out, we observed no differences in VHL30 degradation by hetero-PROTAC 9; e.g., in cIAP1 knockout cells, VHL30 degradation could also be mediated by cIAP2 (Figure S12).A set of experiments were performed to demonstrate the involvement of the ubiquitin−proteasome system in degradation.Treatment of cells with a proteasome inhibitor MG132 completely abrogated degradation of IAPs.The reliance on CRL2 VHL was assessed with a neddylation inhibitor MLN4924, which blocks the activity of CRLs (Figure S13A).Similarly, a selective ubiquitin-activating enzyme inhibitor MLN7243 also prevented the PROTAC-induced degradation of IAPs (Figure S13A).
Concentration-and time-dependent degradation of IAPs in MM.1S cells was evaluated for 25 too (Figures S14 and S10B).Complete cIAP1 and XIAP depletion occurred already at 10 nM, whereas 0.1 μM concentration was needed for the complete depletion of all IAPs.The corresponding CRBNnon-binding control 28 failed to degrade XIAP at 1 μM concentration but caused a significant deficit of cIAP2 (32% remaining protein, Figure S14).Pan-depletion of IAPs by 25 was also very rapid as we observed complete degradation already after 3 h at 0.1 μM (Figure S10B).In addition, the proteasome-mediated mechanism of IAP degradation by 25 was confirmed using the same experiments as for hetero-PROTAC 9 (Figure S13B).
To analyze the proteome-wide degradation selectivity of hetero-PROTACs 9, 25, and 27, a diaPASEF-based mass spectrometry approach was employed. 54MM.1S cells were treated with 100 nM PROTACs for 3 h.Of the total 7170 unique proteins identified, 9 (Figure 6A) and 25 (Figure S15) degraded cIAP1 and XIAP to levels below the detection level,  whereas cIAP2 could not be evaluated in this experiment as it was undetected in DMSO−vehicle treatments.Accordingly, global proteomic plots show the mathematically imputed levels of IAP proteins in treatment groups if the corresponding IAP was detected in the control treatment (see also Experimental Section).AZD5582 also depleted cIAP1 below the detection level but did not cause XIAP degradation (Figure 6C), which is in accordance with the fact that IAP antagonists have no effect on XIAP.
In a separate experiment, global proteome analysis in MM.1S cells after treatment with 27 (Figure 6B) showed selective XIAP degradation with no impact on cIAP1 and cIAP2, rendering this compound significantly more selective for XIAP over cIAP1 and cIAP2 that were only degraded at higher concentrations and after prolonged treatment times.Moreover, we did not observe changes in CRBN, IKZF1, IKZF3, and VHL levels in the proteomic data, thus further substantiating the unilateral effect of our hetero-PROTACs.
pan-IAP Degradation Reduces Cell Viability.To assess the pharmacological consequences of IAP depletion, the pan-IAP degraders 9 and 25, along with appropriate inactivated PROTACs, were evaluated for their cell growth inhibition in nine hematological cell lines (Figures 7 and S17), i.e., three myeloma (RPMI-8226, JJN3, and NCI-H929), three leukemia (HEL, K562, and MOLM13), and three lymphoma cells (SUDHL4, DB, and SUDHL6).We included the monovalent ligase ligands VH298 (VHL), pomalidomide (CRBN), and CST530 (IAP), as well as the homobivalent SMAC mimetic AZD5582 as reference standards.As TNF-α and related signaling cascades represent crucial factors for the single-agent activity of IAP-targeting compounds, 29,38,39,55 the viability inhibition of sensitive cell lines was evaluated in the presence and absence of this inflammatory stimulus.Co-administration of TNF-α potentiated the inhibitory effects of both SMAC mimetics and PROTACs for all cell lines tested, consistent with previous studies. 56−59 pan-IAP degraders 9 and 25 were more potent than the positive control IAP monovalent and bivalent antagonists in several cell lines (Figures 7 and S17).PROTACs 9 and 25 demonstrated superior activity over AZD5582 in NCI-H929, reaching IC 50 values of 8.5 and 27 nM, respectively (Tables 1 and S8).In addition, PROTAC 9 demonstrated a competitive IC 50 profile in MOLM13 cells at 2.1 nM and SUDHL6 cells at 1.6 nM.While the activity of PROTAC 25 in these two cell lines did not supersede PROTAC 9, it was able to induce potent cell viability reduction in other cell lines such as JJN3 and SUDHL4 (Table S8), surpassing that of IAP antagonists.These effects were independent of IAP baseline levels (Figure S16), which is in agreement with the previous studies of IAP antagonists.

■ CONCLUSIONS
In this study, we designed heterobifunctional compounds assembled from an IAP antagonist linked to either a VHL-or a CRBN-recruiting ligand.The entire set of PROTACs consisted of 32 tailored members, which were subjected to in-depth biological studies.Through appropriate control experiments (chemical controls and impairment of the ubiquitin− proteasome system), we provided significant evidence for the engagement with the proposed E3 ligases and PROTACinduced ubiquitin transfer.The accompanied heterodimerization approach led to novel E3 modulators with IAP degradation profiles that could not be reached with monomeric or homobivalent SMAC mimetics.Among the set of IAP degraders were compounds that induced depletion of the 19 and 30 kDa VHL isoforms.The described pan-IAP degraders will serve as selective tools to explore the biology of IAPs and thus open up new avenues for apoptosis research in various cellular contexts.In addition, selected compounds from our  series warrant further appraisal as anti-cancer agents on account of their ability of depleting validated cancer-related IAPs.Preliminary cell-based evaluations of our lead hetero-PROTAC 9 demonstrated that induced degradation of IAPs supersedes the biological effects of monovalent and bivalent IAP antagonists in certain cases.Therefore, further development of IAP-targeting heterobifunctional compounds may lead to degraders with significant therapeutic benefits in the battle against cancer.Exploiting PROTAC methodology to induce the degradation of therapeutically relevant ligases raises hope to unlock this difficult-to-tackle class of drug targets.We also anticipate that the contest between two different E3s may be extendable to any other ligandable ligase.
■ EXPERIMENTAL SECTION Chemistry General Remarks.Preparative column chromatography was performed using Merck silica gel 60 (0.063−0.200 mm) or using an automated flash chromatography system puriFlash XS 520Plus.Melting points were determined on a Buchi 510 oil bath apparatus or on a Reichelt hot-stage apparatus and were uncorrected.H NMR and 13 C NMR spectra were recorded on a Bruker Avance 400 MHz NMR spectrometer, a Bruker Avance 500 MHz NMR spectrometer, or a Bruker Avance III 600 MHz NMR spectrometer, respectively.NMR spectra were processed and analyzed in MestReNova.Chemical shifts are given in parts per million (ppm) and are referenced to the deuterated solvent used.Coupling constants J are given in Hz, and the splitting patterns are given as s (singlet), d (doublet), t (triplet), q (quartet), or m (multiplet).In the case of overlapping extraneous solvent peaks, multiplet analyses in 1 H NMR spectra were performed using qGSD (quantitative Global Spectral Deconvolution).Resonance assignments were made on the basis of one-and two-dimensional NMR techniques which include 1 H, 13 C, DEPT, HSQC, and HMBC experiments.Important note: the presence of amide rotamers significantly complicated the appearance and validation of the 1 H and 13 C NMR spectra associated with synthetic intermediates and final PROTACs.The presence of rotamers was demonstrated by recording a representative 1 H NMR at 80 °C (see the Supporting Information).Thus, reported resonances and integrals may have limited accuracy.High-resolution mass measurements were recorded on a Thermo Scientific Q Exactive Plus mass spectrometer (Thermo Fisher Scientific).The purity and identity of compounds were determined on an Infinity Lab LC/MSD system (Agilent) with the ESI source coupled with an HPLC 1260 Infinity II (Agilent) using an EC50/2 Nucleodur C18 Gravity 3 μm column (Macherey-Nagel).The column temperature was 40 °C.HPLC conditions started with 90% H 2 O containing 2 mM NH 4 Ac.The gradient ramped up to 100% MeCN in 10 min, followed by further flushing with 100% MeCN for 5 min.The flow rate was 0.5 mL/min.The samples were dissolved in Figure 7. Cell viability screenings in nine different hematological cancer cell lines with pan-IAP degrader 9, its VHL non-binding control 11, as well as the structurally related IAP antagonist CST530 and the VHL inhibitor VH298 as respective controls.In certain cases, viability inhibition was assessed in the presence and absence of TNF-α.Multiple myeloma, acute myeloid leukemia, and lymphoma cell lines were treated with the respective compounds at indicated concentrations for 96 h.Viability is normalized to their respective DMSO controls.Data represent means ± s.d. of at least three independent biological replicates.
H 2 O, MeOH, or MeCN (approx. 1 mg/mL), and 2 μL of the sample solution was injected.Positive total ion scans were observed from 100 to 1000 m/z (or more if necessary), and UV absorption was detected from 190 to 600 nm using a diode array detector (DAD).The purity was determined at 220−600 nm.Analytical reversed-phase HPLC for PROTACs 11 and 20−27 was performed on a Thermo Scientific Dionex UltiMate 3000 UHPLC modular system (Thermo Fisher Scientific), equipped with a photodiode array detector set to 254 nm.A Waters Acquity UPLC HSS C18 SB column (1.8 μm, 2.1 mm × 50 mm) was used and thermostated at 40 °C.The mobile phase consisted of 0.1% TFA in H 2 O (A) and MeCN (B), employing the following gradient: 95% A to 5% A in 10 min, then 95% B for 4 min, with a flow rate of 0.3 mL/min, and an injection volume of 5 μL.All compounds that were evaluated in biological assays are >95% pure by HPLC analysis.Note: To provide readers a clearer picture of all synthesized compounds and to enable easier tracking of experimental procedures, a table with structures of all intermediates is given at the end of the Supporting Information.
General Procedures.General Procedure I: Mesylation.To a solution of the corresponding alcohol (7 mmol) in dry CH 2 Cl 2 (15 mL), DIPEA (1.36 g, 1.79 mL, 10.5 mmol) was added under an argon atmosphere and the mixture was cooled to 0 °C.Subsequently, methanesulfonyl chloride (1.20 g, 0.81 mL, 10.5 mmol) was added dropwise at 0 °C, followed by stirring of the mixture at rt for 2 h.After the reaction was complete (monitored by TLC), MeOH (20 mL) was added to the mixture carefully.The volatiles were then evaporated, and the crude product was purified by column chromatography.
General Procedure II: Alkylation of the IAP Ligand Using Cl-Bearing Linkers or Conjugates.The corresponding linker or the VHL ligand−linker conjugate (0.30 mmol) was dissolved in dry acetone (15 mL), and NaI (0.45 g, 3.0 mmol) was added.The mixture was stirred at 60 °C for 48 h.After evaporation of the solvent, the residue was suspended in EtOAc (50 mL) and subsequently washed with 10% Na 2 SO 3 solution, H 2 O, and brine (each 25 mL).The organic layer was dried over Na 2 SO 4 , filtered, and evaporated.This intermediate was dissolved in dry DMF (5 mL), and Cs 2 CO 3 and the corresponding IAP ligand 65 or 66 (1.0 equiv based on the yield from the Finkelstein reaction) were added.The mixture was stirred at 60 °C for 18 h.After cooling, it was quenched with halfsaturated brine (100 mL) and extracted with CH 2 Cl 2 (3 × 50 mL).The combined organic layers were washed with 5% LiCl solution and brine (each 50 mL), dried over Na 2 SO 4 , filtered, and evaporated.The crude product was purified by column or flash chromatography.
General Procedure IV: Alkylation of the VHL2 Ligand Using OMs-to-Cl Linkers.The corresponding OMs-to-Cl linker L1b−L8b (1.2 mmol) was dissolved in dry DMF (10 mL), followed by the addition of Cs 2 CO 3 (0.49g, 1.5 mmol).Then, the phenolic VHL ligand 71 (0.55 g, 1.0 mmol) dissolved in dry DMF (5 mL) was added.The combined mixture was stirred at room temperature for 16 h and 3 h at 60 °C.After cooling, half-saturated brine (50 mL) was added, and the product was extracted with EtOAc (3 × 50 mL).The combined organic phases were washed with saturated NH 4 Cl solution, 5% LiCl solution, and brine (each 50 mL); dried over Na 2 SO 4 , filtered, and concentrated in vacuo.The crude product was purified by column or flash chromatography.
General Procedure V: Alkylation of the IAP Ligand Using OMsto-Cl Linkers.To a solution of IAP ligand 65 (0.17 g, 0.25 mmol) and K 2 CO 3 (52 mg, 0.38 mmol) in dry DMF (2 mL), a solution of the corresponding mesylate-bearing linker L1b−L8b (0.30 mmol) in dry DMF (2 mL) was added under an argon atmosphere.The mixture was stirred at 70 °C for 20 h.The volatiles were then evaporated, and the crude product was purified by column chromatography.
General Procedure VI: Synthesis of Alkyl Azides and Subsequent Reduction to Amines.To a solution of the corresponding IAP ligand−linker−chloro conjugate 91−98 (0.19 mmol) in dry DMF (5 mL), NaN 3 (25 mg, 0.38 mmol) was added under an argon atmosphere.After stirring the mixture at 80 °C for 4 h, the volatiles were removed, and H 2 O (40 mL) was added.The product was extracted with EtOAc (60 mL).The organic layer was washed with brine (50 mL), dried over Na 2 SO 4 , filtered, concentrated, and further dried under high vacuum.This azide intermediate was dissolved in dry MeOH (5 mL) and treated with 10% Pd/C (22 mg, 20% w/w).The reaction mixture was stirred under H 2 (1 atm, balloon) for 2 h.The mixture was filtered through Celite and washed with MeOH, and the filtrate was concentrated.The products were used in the next step without further purification.
General Procedure VII: Removal of Boc Protecting Groups.The Boc-protected PROTAC precursor was treated with 1 M HCl in EtOAc (5 mL), and the mixture was stirred at rt for 4 h.After removal of the volatiles, the oily residue was treated with Et 2 O (5 mL), and the mixture was stirred at rt for 1 h.If a colorless precipitate appeared, it was collected by suction filtration and washed with Et 2 O (2 × 2 mL).Because of sufficient purity, PROTACs 2, 4, 6, 9, 10, and 12−17 were used as hydrochloride salts.For the remaining final PROTACs, additional purification by column chromatography was necessary, and those compounds were transformed into free bases.

Figure 2 .
Figure 2. Degradation profiles of Series 1 (left), Series 2 (middle), and Series 3 (right) of hetero-PROTACs on cIAP1, cIAP2, XIAP, VHL30, CRBN, and IKZF3 expression levels.Percentage degradation is indicated as the remaining protein levels after MM.1S cells were subjected to 16 h treatment with each compound at 0.1 μM.Values are normalized to respective loading controls and to DMSO-treated conditions.All data represent an average of at least three independent experiments.CST530: IAP ligand and VH298: VHL ligand.

Figure 6 .
Figure 6.diaPASEF quantitative proteomics for (A) hetero-PROTAC 9 (CST626), (B) hetero-PROTAC 27 (SAB142), and (C) homobivalent compound AZD5582.MM.1S cells were treated with either DMSO or the mentioned compounds at 0.1 μM for 3 h in four and two biological replicates, respectively.Downstream statistical analysis was performed using Bioconductor's limma package.The quantified proteins were plotted as log 2-fold change (compound/DMSO) versus −log 10 of p-value using RStudio.Note: dataset for 27 was obtained in an independent/separate proteomics run.